AN APPLICATION OF INTEGRATED CLUSTERING TO MRI SEGMENTATION

Magnetic Resonance Imaging (MRI) plays a relevant role in the design of systems for computer assisted diagnosis. MR-images are multi-dimensional in nature; radiologists have to combine several perceptual information coming from more than one image (usually less-than-or-equal-to 4). This to perform the tissue classification needed for diagnosis. Automatic clustering methods help to discriminate relevant features and to perform preliminary segmentation of an image; it can guide the manual classification of body-tissues. Here four clustering techniques and their integration, in an information fusion clustering procedure, are described. The accuracy of the methodology considered is evaluated on real image data. The evaluation is based on the comparison of the results obtained by automatic procedures with the tissue-classification done by radiologists.